Pemi Nguyen

Machine Learning Engineer tại Meta

Theo dõi

Hồ sơ
Đánh giá

Chương trình mentoring của Pemi Nguyen

Giới thiệu bản thân

I'm an educator and practitioner in Machine Learning who is passionate about delivering a high-quality computing education with a focus on inclusivity and accessibility to students of all backgrounds, especially those coming from underrepresented groups. As someone born with a disability, I would like to provide dedicated support to anyone who is facing barriers in entering the tech field. I'm also interested in giving industry perspectives to excite people about the applications of Machine Learning in practice.

As an avid champion of many societal causes (youth empowerment, environmental protection, disability inclusion, etc.), I have attended many conferences all over the world and received trainings sponsored by established organizations such as One Young World, U.S. Department of State's Young Southeast Asian Leaders Initiative (YSEALI), Asia-Europe Foundation (ASEF), Asians and Pacific Islanders with Disabilities of California (APIDC).

At the moment, as a machine learning engineer in the industry, I am currently interested in exploring the use of machine learning in production (deployment tools, distributed systems), enhancing existing models and datasets that deliver high-quality business results and great social impact, and implementing state-of-the-art architectures from academic literature.

Kinh nghiệm làm việc

Machine Learning Engineer

Meta

Recommendation Ecosystem ML team

06/2022 - Hiện tại
Paul G. Allen School of Computer Science & Engineering

Paul G. Allen School

* Feb 2022 - Jun 2022: Lecturer for CSE 416 (Introduction to Machine Learning):
- Taught introductory Machine Learning to a class of 160 advanced undergrad and graduate students
- Managed a staff team of 10 Teaching Assistants (5 undergrads and 5 grads)
- Developed teaching materials on a number of topics such as clustering, nearest neighbor classification using various distance metrics, PCA, word embeddings, transfer learning, machine learning biases, etc.
- Implemented autograders for coding assignments to increase grading efficiency

* Sep 2020 - Mar 2022: NLP Research Assistant - Yejin Choi's Lab
· Fine-tune BERT on a dataset consisting of 51,000 propagandist and non-propagandist articles to detect whether a certain article contains propagandistic content or not, which achieves a test accuracy of 97.41%.
· Deploy the detection model and build a web scraping tool on a system where users can check online articles’ reliability
· Examine articles that are incorrectly predicted to determine what text patterns can make the models fail

* Dec 2020 - Mar 2022: TA and Summer 2021 Project Leader - CSE 446/546 (Machine Learning)
· Prepare assignments and materials which allows students to practice theoretical concepts and Python coding on multiple topics: bias-variance tradeoff, convexity, gradient descent, kernels, neural networks, deep learning
· Lead a team of 2 TAs that work on adding new problems, quizzes and topics that are up-to-date with the ML field
· Contribute to the codebase, infrastructure and unit tests for coding assignments with detailed documentations

12/2020 - 06/2022

Quá trình học tập

ASEFEdu 21st Summer University

Disability Study

The ASEF Summer University (ASEFSU) is a 2-week experiential learning journey and “Interdisciplinary Innovathon” for Asian and European students and young professionals. Designed to foster cross-cultural exchanges and networks among youth, it offers opportunities for students and young professionals to broaden their horizons, deepen their knowledge on contemporary issues, and propose concrete solutions to societal challenges.

It was conducted under the theme "Youth with Disabilities: Shaping Inclusive ASEM Societies" and it consists of a mix of practical team exercises, thought-provoking lectures, group research and experiential learning. Participants will develop a strong understanding for the issues faced by youth with disabilities and become active ambassadors for open and inclusive ASEM societies.
University of Washington - Paul G. Allen School of Computer Science & Engineering

Computer Science

Hoạt động ngoại khóa

Reproduction of the “Dense Passage Retrieval” EMNLP 2020 Paper

Project Leader

This is a project that attempts to reproduce and verify the main claims of "Dense Passage Retrieval for Open-Domain Question Answering" (Karpukhin et al., EMNLP 2020) with the final report following the format of submissions to the ML Reproducibility Challenge. This project was done as a final project for CSE 517 wi21, taught by Prof. Noah Smith.

Tên giải thưởng

Publications

- Introductory Machine Learning Textbook
This is the supplementary textbook for UW CSE 416, Introduction to Machine Learning.

- Misinfo Reaction Frames: Reasoning about Readers' Reactions to News Headlines
Prior beliefs of readers impact the way in which they project meaning onto news headlines. These beliefs can influence their perception of news reliability, as well as their reaction to news, and their likelihood of spreading the misinformation through social networks. However, most prior work focuses on fact-checking veracity of news or stylometry rather than measuring impact of misinformation. We propose Misinfo Belief Frames, a formalism for understanding how readers perceive the reliability of news and the impact of misinformation. We also introduce the Misinfo Belief Frames (MBF) corpus, a dataset of 66k inferences over 23.5k headlines. Misinformation frames use commonsense reasoning to uncover implications of real and fake news headlines focused on global crises: the Covid-19 pandemic and climate change. Our results using large-scale language modeling to predict misinformation frames show that machine-generated inferences can influence readers' trust in news headlines (readers' trust in news headlines was affected in 29.3% of cases). This demonstrates the potential effectiveness of using generated frames to counter misinformation.

- Probability and Statistics for Computer Science Textbook
This is the official textbook for the course CSE 312 : Probability and Statistics for Computer Science at the University of Washington.